4 ms·
At the risk of being overly cynical: Week 1: Take 512 people. Mail half of them a letter that says "the market will go down this week". Mail the other half a l
by numlocked 12y ago
At the risk of being overly cynical:
Week 1: Take 512 people. Mail half of them a letter that says "the market will go down this week". Mail the other half a letter claiming it will go up.
Week 2: Take the half who received a correct prediction, split into 2 groups, repeat.
Week 3: Repeat.
Week 4: Repeat.
Week 5: "I've predicted market movements perfectly! To get on board with my foolproof strategy, send your investment check to..."
- im2w1l 12y agoThis seems to be the opposite. Week 1: 512 participants. Half of them mail in "the market will go down this week". They lose. Week 2: Of those who made a correct prediction, half make a correct prediction again. The rest lose. Week 3-8. Etc Week 9. We found a winner, one who could predict the market flawlessly!
- fawce 12y agoTo be fair, we invested our own money in the winner. The composition of the algo portfolio for the fund != the winners only. Our challenge is to choose the optimal set of algorithms from a huge pool. Optimal means maximizing returns while also minimizing correlation between the strategies. The holy grail of investing is 10 or so uncorrelated return streams. We're building a community and platform to consistently turn out algorithms with uncorrelated return streams.